Evaluating the different parameters in the model
We are imputing our data for a reason - we want to analyze the data!
In this example, we are interested in predicting sea temperature, so we will build a linear model predicting sea temperature.
We will fit this model to each of the datasets we created and then explore the coefficients in the data.
The objects from the previous lesson (ocean_cc, ocean_imp_lm_wind, ocean_imp_lm_all, and bound_models) are loaded into the workspace.
This exercise is part of the course
Dealing With Missing Data in R
Exercise instructions
- Create the model summary for each dataset with columns for residuals using
residuals,predict, andtidy. - Explore the coefficients in the model and put the model with the highest estimate for
air_temp_cin the object best_model
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Create the model summary for each dataset
model_summary <- bound_models %>%
group_by(imp_model) %>%
nest() %>%
mutate(mod = map(data, ~lm(sea_temp_c ~ air_temp_c + humidity + year, data = .)),
res = map(mod, ___),
pred = map(mod, ___),
tidy = map(mod, ___))
# Explore the coefficients in the model
model_summary %>%
select(___,___) %>%
unnest()
best_model <- "___"